prjct-cli 1.22.0 → 1.24.0

package/templates/global/modules/CLAUDE-commands.md

@@ -1,70 +0,0 @@
-## FAST vs SMART COMMANDS (CRITICAL)
-
-**Some commands just run a CLI. Others need intelligence. Know the difference.**
-
-### FAST COMMANDS (Execute Immediately - NO planning, NO exploration)
-
-| Command | Action | Time |
-|---------|--------|------|
-| `p. sync` | Run `prjct sync` | <5s |
-| `p. next` | Run `prjct next` | <2s |
-| `p. dash` | Run `prjct dash` | <2s |
-| `p. pause` | Run `prjct pause` | <2s |
-| `p. resume` | Run `prjct resume` | <2s |
-
-**For these commands:**
-```
-1. Read template
-2. Run the CLI command shown
-3. Done
-```
-
-**DO NOT:** explore codebase, create plans, ask questions, read project files
-
-### SMART COMMANDS (Require intelligence)
-
-| Command | Why it needs intelligence |
-|---------|--------------------------|
-| `p. task` | Must explore codebase, break down work |
-| `p. ship` | Must validate changes, create PR |
-| `p. bug` | Must classify severity, find affected files |
-| `p. done` | Must verify completion, update state |
-
-**For these commands:** Follow the full INTELLIGENT BEHAVIOR rules.
-
-### Decision Rule
-```
-IF template just says "run CLI command":
-  → Execute immediately, no planning
-ELSE:
-  → Use intelligent behavior (explore, ask, plan)
-```
-
----
-
-## CORE WORKFLOW
-
-```
-p. sync  →  p. task "description"  →  [work]  →  p. done  →  p. ship
-   │              │                                  │            │
-   │              └─ Creates branch, breaks down     │            │
-   │                 task, starts tracking           │            │
-   │                                                 │            │
-   └─ Analyzes project, generates agents             │            │
-                                                     │            │
-                              Completes subtask ─────┘            │
-                                                                  │
-                                        Ships feature, PR, tag ───┘
-```
-
-### Quick Reference
-
-| Trigger | What It Does |
-|---------|--------------|
-| `p. sync` | Analyze project, generate domain agents |
-| `p. task <desc>` | Start task with auto-classification |
-| `p. done` | Complete current subtask |
-| `p. ship [name]` | Ship feature with PR + version bump |
-| `p. pause` | Pause current task |
-| `p. resume` | Resume paused task |
-| `p. bug <desc>` | Report bug with auto-priority |

package/templates/global/modules/CLAUDE-core.md

@@ -1,105 +0,0 @@
-# prjct-cli
-
-**Context layer for AI agents** - Project context for Claude Code, Gemini CLI, and more.
-
-## HOW TO USE PRJCT
-
-When user types `p. <command>`, **READ the template** from `~/.claude/commands/p/{command}.md` and execute it step by step.
-
-```
-p. sync     → ~/.claude/commands/p/sync.md
-p. task X   → ~/.claude/commands/p/task.md
-p. done     → ~/.claude/commands/p/done.md
-p. ship X   → ~/.claude/commands/p/ship.md
-```
-
-**⚠️ ALWAYS Read() the template file first. Templates contain mandatory workflows.**
-
----
-
-## LOADING DOMAIN AGENTS (CRITICAL)
-
-**Before starting any 🧠 SMART command (task, ship, bug, done):**
-
-```
-1. Read .prjct/prjct.config.json → get projectId
-2. Set globalPath = ~/.prjct-cli/projects/{projectId}
-3. Read {globalPath}/agents/*.md for domain expertise:
-   - prjct-planner.md → for task planning
-   - prjct-shipper.md → for shipping
-   - backend.md, frontend.md, etc → for domain-specific work
-```
-
-**USE the agent context when working.** Agents contain project-specific patterns.
-
----
-
-## CRITICAL RULES
-
-### 0. FOLLOW TEMPLATES STEP BY STEP (NON-NEGOTIABLE)
-
-**Templates are MANDATORY WORKFLOWS, not suggestions.**
-
-```
-1. READ the template file COMPLETELY
-2. FOLLOW each step IN ORDER
-3. DO NOT skip steps - even "obvious" ones
-4. STOP at any BLOCKING condition
-```
-
-### 1. Path Resolution (MOST IMPORTANT)
-
-**ALL writes go to global storage**: `~/.prjct-cli/projects/{projectId}/`
-
-- **NEVER** write to `.prjct/` (config only, read-only)
-- **NEVER** write to `./` (current directory)
-- **ALWAYS** resolve projectId first from `.prjct/prjct.config.json`
-
-### 2. Before Any Command
-
-```
-1. Read .prjct/prjct.config.json → get projectId
-2. Set globalPath = ~/.prjct-cli/projects/{projectId}
-3. Execute command using globalPath for all writes
-4. Log to {globalPath}/memory/events.jsonl
-```
-
-### 3. Timestamps & UUIDs
-
-```bash
-# Timestamp (NEVER hardcode)
-bun -e "console.log(new Date().toISOString())" 2>/dev/null || node -e "console.log(new Date().toISOString())"
-
-# UUID
-bun -e "console.log(crypto.randomUUID())" 2>/dev/null || node -e "console.log(require('crypto').randomUUID())"
-```
-
----
-
-## OUTPUT FORMAT
-
-Concise responses (< 4 lines):
-```
-[What was done]
-
-[Key metrics]
-Next: [suggested action]
-```
-
----
-
-## CLEAN TERMINAL UX
-
-**Tool calls MUST be user-friendly.**
-
-1. **ALWAYS use clear descriptions** in Bash tool calls:
-   - GOOD: `description: "Building project"`
-   - BAD: `description: "bun run build 2>&1 | tail -5"`
-
-2. **Hide implementation details** - Users don't need to see pipe chains, internal paths, JSON parsing
-
-3. **Use action verbs**: "Building project", "Running tests", "Checking git status"
-
----
-
-**Auto-managed by prjct-cli** | https://prjct.app

package/templates/global/modules/CLAUDE-git.md

@@ -1,50 +0,0 @@
-## GIT WORKFLOW RULES (CRITICAL)
-
-**NEVER commit directly to main/master**
-- Always create a feature branch first
-- Always create a PR for review
-- Direct pushes to main are FORBIDDEN
-
-**NEVER push without a PR**
-- All changes go through pull requests
-- No exceptions for "small fixes"
-
-**NEVER skip version bump on ship**
-- Every ship requires version update
-- Every ship requires CHANGELOG entry
-
-### Git Commit Footer (CRITICAL - ALWAYS INCLUDE)
-
-**Every commit made with prjct MUST include this footer:**
-
-```
-Generated with [p/](https://www.prjct.app/)
-```
-
-**This is NON-NEGOTIABLE. The prjct signature must appear in ALL commits.**
-
-### PLAN BEFORE DESTRUCTIVE ACTIONS
-
-For commands that modify git state (ship, merge, done):
-```
-1. Show the user what will happen
-2. List all changes/files affected
-3. WAIT for explicit approval ("yes", "proceed", "do it")
-4. Only then execute
-```
-
-**DO NOT assume approval. WAIT for it.**
-
-### BLOCKING CONDITIONS
-
-When a template says "STOP" or has a blocking symbol:
-```
-1. HALT execution immediately
-2. TELL the user why you stopped
-3. DO NOT proceed until the condition is resolved
-```
-
-**Examples of blockers:**
-- `p. ship` on main branch → STOP, tell user to create branch
-- `gh auth status` fails → STOP, tell user to authenticate
-- No changes to commit → STOP, tell user nothing to ship

package/templates/global/modules/CLAUDE-intelligence.md

@@ -1,92 +0,0 @@
-## INTELLIGENT BEHAVIOR (For SMART commands only)
-
-### When Starting Tasks (`p. task`)
-1. **Analyze** - Understand what user wants to achieve
-2. **Classify** - Determine type: feature, bug, improvement, refactor, chore
-3. **Explore** - Find similar code, patterns, affected files
-4. **Ask** - Clarify ambiguities (use AskUserQuestion)
-5. **Design** - Propose 2-3 approaches, get approval
-6. **Break down** - Create actionable subtasks
-7. **Track** - Update storage/state.json
-
-### When Completing Tasks (`p. done`)
-1. Check if there are more subtasks
-2. If yes, advance to next subtask
-3. If no, task is complete
-4. Update storage, generate context
-
-### When Shipping (`p. ship`)
-1. Run tests (if configured)
-2. Create PR (if on feature branch)
-3. Bump version
-4. Update CHANGELOG
-5. Create git tag
-
-### Key Intelligence Rules
-- **Read before write** - Always read existing files before modifying
-- **Explore before coding** - Use Task(Explore) to understand codebase
-- **Ask when uncertain** - Use AskUserQuestion to clarify
-- **Log everything** - Append to memory/events.jsonl
-
----
-
-## ARCHITECTURE: Write-Through Pattern
-
-```
-User Action → Storage (JSON) → Context (MD) → Sync Events
-```
-
-| Layer | Path | Purpose |
-|-------|------|---------|
-| **Storage** | `storage/*.json` | Source of truth |
-| **Context** | `context/*.md` | Claude-readable summaries |
-| **Memory** | `memory/events.jsonl` | Audit trail (append-only) |
-| **Agents** | `agents/*.md` | Domain specialists |
-| **Sync** | `sync/pending.json` | Backend sync queue |
-
-### File Structure
-```
-~/.prjct-cli/projects/{projectId}/
-├── storage/
-│   ├── state.json      # Current task (SOURCE OF TRUTH)
-│   ├── queue.json      # Task queue
-│   └── shipped.json    # Shipped features
-├── context/
-│   ├── now.md          # Current task (generated)
-│   └── next.md         # Queue (generated)
-├── memory/
-│   └── events.jsonl    # Audit trail
-├── agents/             # Domain specialists (auto-generated)
-└── sync/
-    └── pending.json    # Events for backend
-```
-
----
-
-## LOADING DOMAIN AGENTS
-
-When working on tasks, load relevant agents from `{globalPath}/agents/`:
-- `frontend.md` - Frontend patterns, components
-- `backend.md` - Backend patterns, APIs
-- `database.md` - Database patterns, queries
-- `uxui.md` - UX/UI guidelines
-- `testing.md` - Testing patterns
-- `devops.md` - CI/CD, containers
-
-These agents contain project-specific patterns. **USE THEM**.
-
----
-
-## SKILL INTEGRATION
-
-Agents are linked to Claude Code skills from claude-plugins.dev.
-
-### How Skills Work
-
-1. **During `p. sync`**: Search claude-plugins.dev, install best matches
-2. **During `p. task`**: Skills are auto-invoked for domain expertise
-3. **Agent frontmatter** has `skills: [discovered-skill-name]` field
-
-### Skill Location
-
-Skills are markdown files in `~/.claude/skills/`

package/templates/global/modules/CLAUDE-storage.md

@@ -1,50 +0,0 @@
-## STORAGE RULES (CROSS-AGENT COMPATIBILITY)
-
-**NEVER use temporary files** - Write directly to final destination:
-- WRONG: Create `.tmp/file.json`, then `mv` to final path
-- CORRECT: Write directly to `{globalPath}/storage/state.json`
-
-**JSON formatting** - Always use consistent format:
-- 2-space indentation
-- No trailing commas
-- Keys in logical order (as defined in storage schemas)
-
-**Atomic writes for JSON**:
-```javascript
-// Read → Modify → Write (no temp files)
-const data = JSON.parse(fs.readFileSync(path, 'utf-8'))
-data.newField = value
-fs.writeFileSync(path, JSON.stringify(data, null, 2))
-```
-
-**Timestamps**: Always ISO-8601 with milliseconds (`.000Z`)
-**UUIDs**: Always v4 format (lowercase)
-**Line endings**: LF (not CRLF)
-**Encoding**: UTF-8 without BOM
-
-**NEVER**:
-- Use `.tmp/` directories
-- Use `mv` or `rename` operations for storage files
-- Create backup files like `*.bak` or `*.old`
-- Modify existing lines in `events.jsonl`
-
-**Full specification**: See `{npm root -g}/prjct-cli/templates/global/STORAGE-SPEC.md`
-
----
-
-## Preserve Markers (User Customizations)
-
-User customizations in context files and agents survive regeneration using preserve markers:
-
-```markdown
-<!-- prjct:preserve -->
-# My Custom Rules
-- Always use tabs
-- Prefer functional patterns
-<!-- /prjct:preserve -->
-```
-
-**How it works:**
-- Content between markers is extracted before regeneration
-- After regeneration, preserved content is appended under "Your Customizations"
-- Named sections: `<!-- prjct:preserve:my-rules -->` for identification

package/templates/global/modules/module-config.json

@@ -1,36 +0,0 @@
-{
-  "description": "Configuration for modular CLAUDE.md composition",
-  "version": "1.0.0",
-  "profiles": {
-    "full": {
-      "description": "All modules - maximum context (~2300 tokens)",
-      "modules": [
-        "CLAUDE-core.md",
-        "CLAUDE-commands.md",
-        "CLAUDE-git.md",
-        "CLAUDE-storage.md",
-        "CLAUDE-intelligence.md"
-      ]
-    },
-    "standard": {
-      "description": "Standard modules - balanced (~1400 tokens)",
-      "modules": ["CLAUDE-core.md", "CLAUDE-commands.md", "CLAUDE-git.md"]
-    },
-    "minimal": {
-      "description": "Core only - minimum context (~500 tokens)",
-      "modules": ["CLAUDE-core.md"]
-    }
-  },
-  "default": "standard",
-  "commandProfiles": {
-    "sync": "minimal",
-    "next": "minimal",
-    "dash": "minimal",
-    "pause": "minimal",
-    "resume": "minimal",
-    "task": "full",
-    "done": "standard",
-    "ship": "full",
-    "bug": "full"
-  }
-}

package/templates/mcp-config.json

@@ -1,19 +0,0 @@
-{
-  "mcpServers": {
-    "context7": {
-      "command": "npx",
-      "args": ["-y", "@upstash/context7-mcp@latest"],
-      "description": "Library documentation lookup"
-    }
-  },
-  "usage": {
-    "context7": {
-      "when": ["Looking up library/framework documentation", "Need current API docs"],
-      "tools": ["resolve-library-id", "get-library-docs"]
-    }
-  },
-  "integrations": {
-    "linear": "SDK - Set LINEAR_API_KEY env var",
-    "jira": "REST API - Set JIRA_BASE_URL, JIRA_EMAIL, JIRA_API_TOKEN env vars"
-  }
-}

package/templates/permissions/default.jsonc

@@ -1,60 +0,0 @@
-{
-  // Default permissions preset for prjct-cli
-  // Safe defaults with protection against destructive operations
-
-  "bash": {
-    // Safe read-only commands - always allowed
-    "git status*": "allow",
-    "git log*": "allow",
-    "git diff*": "allow",
-    "git branch*": "allow",
-    "ls*": "allow",
-    "pwd": "allow",
-    "cat*": "allow",
-    "head*": "allow",
-    "tail*": "allow",
-    "grep*": "allow",
-    "find*": "allow",
-    "which*": "allow",
-    "node -e*": "allow",
-    "bun -e*": "allow",
-    "npm list*": "allow",
-    "npx tsc --noEmit*": "allow",
-
-    // Potentially destructive - ask first
-    "rm -rf*": "ask",
-    "rm -r*": "ask",
-    "git push*": "ask",
-    "git reset --hard*": "ask",
-    "npm publish*": "ask",
-    "chmod*": "ask",
-
-    // Always denied - too dangerous
-    "rm -rf /*": "deny",
-    "rm -rf ~/*": "deny",
-    "sudo*": "deny"
-  },
-
-  "files": {
-    "read": {
-      "**/*": "allow"
-    },
-    "write": {
-      "**/*": "allow"
-    },
-    "delete": {
-      "**/*": "ask"
-    }
-  },
-
-  "web": {
-    "enabled": true
-  },
-
-  "doomLoop": {
-    "enabled": true,
-    "maxRetries": 3
-  },
-
-  "externalDirectories": "ask"
-}

package/templates/permissions/permissive.jsonc

@@ -1,49 +0,0 @@
-{
-  // Permissive preset for prjct-cli
-  // For trusted environments - minimal restrictions
-
-  "bash": {
-    // Most commands allowed
-    "git*": "allow",
-    "npm*": "allow",
-    "bun*": "allow",
-    "node*": "allow",
-    "ls*": "allow",
-    "cat*": "allow",
-    "mkdir*": "allow",
-    "cp*": "allow",
-    "mv*": "allow",
-    "rm*": "allow",
-    "chmod*": "allow",
-
-    // Still protect against catastrophic mistakes
-    "rm -rf /*": "deny",
-    "rm -rf ~/*": "deny",
-    "sudo rm -rf*": "deny",
-    ":(){ :|:& };:*": "deny"
-  },
-
-  "files": {
-    "read": {
-      "**/*": "allow"
-    },
-    "write": {
-      "**/*": "allow"
-    },
-    "delete": {
-      "**/*": "allow",
-      "**/node_modules/**": "deny" // Protect dependencies
-    }
-  },
-
-  "web": {
-    "enabled": true
-  },
-
-  "doomLoop": {
-    "enabled": true,
-    "maxRetries": 5
-  },
-
-  "externalDirectories": "allow"
-}

package/templates/permissions/strict.jsonc

@@ -1,58 +0,0 @@
-{
-  // Strict permissions preset for prjct-cli
-  // Maximum safety - requires approval for most operations
-
-  "bash": {
-    // Only read-only commands allowed
-    "git status": "allow",
-    "git log*": "allow",
-    "git diff*": "allow",
-    "ls*": "allow",
-    "pwd": "allow",
-    "cat*": "allow",
-    "head*": "allow",
-    "tail*": "allow",
-    "which*": "allow",
-
-    // Everything else requires approval
-    "git*": "ask",
-    "npm*": "ask",
-    "bun*": "ask",
-    "node*": "ask",
-    "rm*": "ask",
-    "mv*": "ask",
-    "cp*": "ask",
-    "mkdir*": "ask",
-
-    // Always denied
-    "rm -rf*": "deny",
-    "sudo*": "deny",
-    "chmod 777*": "deny"
-  },
-
-  "files": {
-    "read": {
-      "**/*": "allow",
-      "**/.*": "ask", // Hidden files need approval
-      "**/.env*": "deny" // Never read env files
-    },
-    "write": {
-      "**/*": "ask" // All writes need approval
-    },
-    "delete": {
-      "**/*": "deny" // No deletions without explicit override
-    }
-  },
-
-  "web": {
-    "enabled": true,
-    "blockedDomains": ["localhost", "127.0.0.1", "internal"]
-  },
-
-  "doomLoop": {
-    "enabled": true,
-    "maxRetries": 2
-  },
-
-  "externalDirectories": "deny"
-}